I have a question about a fluid flow problem in a porous media. I have a brick that has very small air gap openings. I am injecting a gas into this brick and the porosity of the brick allows the gas to flow through to the other side. I am wondering if there is a way to get the average velocity of the gas leaving the brick instead of the local gas velocity occuring inside the porous material. Using the momentum equation it states that the gas inside the brick will have high velocities, I am not interested in the high velocity inside the brick. I am interested in the lower average velocity exiting the brick wall. Is there a way to program fluent to find the average smaller velocity exiting the brick?

By default, FLUENT perform porous media calculations using the superficial velocity, which is exactly what you seem to want. While actual velocities in the brick will be higher, the superficial velocity formulation doesn't calculate them.

I am still not confident that we want to find the superficial velocity for this ceramic brick. First of all the openings are very very small.

If I understand this right we basically assume that the fluid is flowing inside a open box that is 100% open for instance. We calculate for example the velocity based on the the cons. of mass and momentum. We then define the parameters which give us a very high velocity exiting the material and just use the relation vphysical=vsuperficial/porosity to calculate the superficial velocity. It seems we are loosing a lot of information about what is taking place inside the material. How accurate is that assumption. If we try to solve for the physical velocity i'm afraid that our material is packed too tight and the mesh we create might not be enough to give us good results either. Could we assume the material is a packed bed. Under what criteria can we assume that? How do we calculated loss coefficients based on the superficial velocity techniques? I would appreciate advice on how to go about attacking this problem and what would probably give the most accurate results.

You are losing a lot of information about what happend inside. THere is an option to use actual velocities in FLUENT 6, which I believe assumes uniform resistance properties in an given flow direction. It is still just a "black box" momentum sink term and should give the same results as the superficial velocity formulation. As for calcuating the loss coefficients, this is explained in the FLUENT User's Manual.